From: David Gibson <david@gibson.dropbear.id.au>
To: Reza Arbab <arbab@linux.ibm.com>
Cc: Daniel Henrique Barboza <danielhb@linux.ibm.com>,
Leonardo Augusto Guimaraes Garcia <lagarcia@linux.ibm.com>,
qemu-ppc@nongnu.org, qemu-devel@nongnu.org,
Greg Kurz <groug@kaod.org>
Subject: Re: [PATCH v3] spapr: Add a new level of NUMA for GPUs
Date: Thu, 16 Jul 2020 15:04:59 +1000 [thread overview]
Message-ID: <20200716050459.GM93134@umbus.fritz.box> (raw)
In-Reply-To: <20200525174927.aky64nw7p7xztqzh@arbab-vm>
[-- Attachment #1: Type: text/plain, Size: 2111 bytes --]
On Mon, May 25, 2020 at 12:49:27PM -0500, Reza Arbab wrote:
> On Mon, May 25, 2020 at 03:05:50PM +1000, David Gibson wrote:
> > On Fri, May 22, 2020 at 02:53:33PM -0500, Reza Arbab wrote:
> > > --- a/hw/ppc/spapr_pci_nvlink2.c
> > > +++ b/hw/ppc/spapr_pci_nvlink2.c
> > > @@ -362,7 +362,7 @@ void spapr_phb_nvgpu_ram_populate_dt(SpaprPhbState *sphb, void *fdt)
> > > uint32_t associativity[] = {
> > > cpu_to_be32(0x4),
> > > SPAPR_GPU_NUMA_ID,
> > > - SPAPR_GPU_NUMA_ID,
> > > + cpu_to_be32(nvslot->numa_id),
> > > SPAPR_GPU_NUMA_ID,
> > > cpu_to_be32(nvslot->numa_id)
> >
> >
> > This doesn't look quite right. In the new case we'll get {
> > GPU_NUMA_ID, nvslot->numa_id, GPU_NUMA_ID, nvslot->numa_id }.
>
> The associativity reference points are 4 (and now 2), so this is what we
> want. I think what you've noticed is that reference points are 1-based
> ordinals:
>
> "...the “ibm,associativity-reference-points” property indicates
> boundaries between associativity domains presented by the
> “ibm,associativity” property containing “near” and “far” resources. The
> first such boundary in the list represents the 1 based ordinal in the
> associativity lists of the most significant boundary, with subsequent
> entries indicating progressively less significant boundaries."
Right.. AIUI, associativity-reference-points indicates which leves are
"important" from a NUMA distance point of view (though the spec is
very confusing). But, I'm pretty sure, that ignoring
reference-points, the individual ibm,associativity lists are supposed
to describe a correct hierarchy, even if some levels will get ignored
for distance purposes. So once you've split up into "numa_id" nodes
at the second level, you can't then go back to just 2 nodes (main
vs. gpu) at the third.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2020-07-16 5:20 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-22 19:53 [PATCH v3] spapr: Add a new level of NUMA for GPUs Reza Arbab
2020-05-22 20:08 ` Reza Arbab
2020-05-25 5:06 ` David Gibson
2020-05-25 5:05 ` David Gibson
2020-05-25 17:49 ` Reza Arbab
2020-07-16 5:04 ` David Gibson [this message]
2020-07-16 9:42 ` Daniel Henrique Barboza
2020-07-16 16:00 ` Reza Arbab
2020-07-16 16:40 ` Daniel Henrique Barboza
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200716050459.GM93134@umbus.fritz.box \
--to=david@gibson.dropbear.id.au \
--cc=arbab@linux.ibm.com \
--cc=danielhb@linux.ibm.com \
--cc=groug@kaod.org \
--cc=lagarcia@linux.ibm.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).